About me

I am a PhD student in the Department of Electrical and Electronic Engineering at the University of Hong Kong, advised by Dr. Yanchao Yang (from the HKU Institute of Data Science) and Prof. Kaibin Huang. Before joining HKU, I received my B.Sc. in Mathematics and Applied Mathematics from the School of the Gifted Young, University of Science and Technology of China (USTC).

My research lies at the intersection of statistical dependence modeling, foundation models for data science, and interpretability of large language models. I am especially interested in building general-purpose, pre-trained models that turn classical statistical questions — such as mutual information estimation and dependency measurement — into fast, zero-shot inferences, and in using such tools to monitor and diagnose the training dynamics of modern Transformers.

I am a long-term collaborator with the HPC group (RAFT) at Huawei, where I work on real-time representation monitoring and interpretability for large-scale LLM training.

Research interests

• Foundation models for statistical inference (mutual information, dependency measurement)
• Training dynamics and interpretability of Transformers / LLMs
• Information theory and representation learning

News

• May 2026 — Invited talk at HKU ECE: Foundation-style Methods for Real-Time Statistical Dependency Measurement and Its Applications.
• 2026 — A Foundation-style Model for Zero-Shot Statistical Dependency Measurement accepted to ICML 2026.
• 2024 — InfoNet: Neural Estimation of Mutual Information without Test-time Optimization accepted to ICML 2024 (Oral).

Contact

Email  /  Google Scholar  /  GitHub  /  CV (PDF)